DWI with FSE

Fast (or ''Turbo'') spin-echo imaging (FSE or TSE) has dramatically altered daily clinical practice as a fast and robust image readout strategy with negligible T2* artifacts. Due to significant RF pulse specific absorption rate (SAR) constraints made worse by variable RF pulse flip angle distributions occurring during a long echo train, FSE has been a poor choice for DWI or DTI. One approach for diffusion-weighted FSE imaging has been the U-FLARE technique (Norris et al., 1992), which relies on the separation of different echo families so that they do not inter

Dwi Chemical Shift Artifact

Fig. 3.54. Comparison of conventional singleshot EPI DWI (top) with the corresponding single-shot SENSE-DWI sequence (bottom) from three diffusion-weighted slices of 20 acquired from a suspected stroke with an accompanying subcortical hemorrhagic transformation. The b = 1000 s mm~2 images are shown. By means of the faster k-space

Fig. 3.54. Comparison of conventional singleshot EPI DWI (top) with the corresponding single-shot SENSE-DWI sequence (bottom) from three diffusion-weighted slices of 20 acquired from a suspected stroke with an accompanying subcortical hemorrhagic transformation. The b = 1000 s mm~2 images are shown. By means of the faster k-space traversal with SENSE (acceleration factor, r = 2), the chemical shift artifact can be strongly reduced (arrows). Moreover, magnetic susceptibility artifacts or artifacts from 60-inhomogeneities can be also markedly diminished (open arrows). Note the improved conspicuity of the hemorrhagic lesion on the SENSE-DWI images.

fere. Le Roux and colleagues (Le Roux, 1998, 2000) presented a valid method for diffusion-weighted FSE-DWI which relies on an FSE employing a modified phase setting of the refocusing RF pulses. This approach has become the basis for the ''PROPELLER'' FSE-DWI method, a novel and effective FSE sequence for DWI described by Pipe et al. (2002).

The PROPELLER (Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction) method, which has inherent 2D navigator information in each FSE echo train, provides far greater immunity to geometric and T2* distortions than that obtained with EPI sequences, while providing robust immunity to motion artifacts by collecting FSE trains in various frequency and phase directions in k-space, called ''blades''. Each blade passes through the center of frequency and phase space, which allows for a very good means of''navigating'' each blade to any phase-inducing motional effects. In subsequent pluse repetition time (TRs) this blade is rotated, so that together the blades measure a circular region of k-space formed by their union. The basic idea of PROPELLER is that the blades can correct for inconsistencies prior to combining the data, so the primary inconsistency will be motion-caused phase differences. After data correction, the blades are combined to form the DWI image. Images from the commonly-used SE-EPI DWI sequence and the corresponding PROPELLER sequence for routine stroke DWI are compared in Figure 3.56.

Sequence Epi

Fig. 3.55. Comparison of SE-EPI DWI and ''Turbo-PROPELLER'' DWI for imaging of stroke and hemorrhage. Shown are the SE-EPI DWI (128 x 128; b = 1000) diffusion-weighted images from four of 28 slices acquired in 24 s. Below are the corresponding Turbo-PROP images (128 x 128; ETL 16; b = 1000) from four of 18 slices acquired in 120 s. Single-shot SE-EPI DWI demonstrates significant susceptibility artifacts near air-tissue interfaces (top row). Typical signal pile up and loss as

Fig. 3.55. Comparison of SE-EPI DWI and ''Turbo-PROPELLER'' DWI for imaging of stroke and hemorrhage. Shown are the SE-EPI DWI (128 x 128; b = 1000) diffusion-weighted images from four of 28 slices acquired in 24 s. Below are the corresponding Turbo-PROP images (128 x 128; ETL 16; b = 1000) from four of 18 slices acquired in 120 s. Single-shot SE-EPI DWI demonstrates significant susceptibility artifacts near air-tissue interfaces (top row). Typical signal pile up and loss as well as strong geometric distortions are apparent, obscuring tissue pathology through most slices (bottom row). Turbo-PROP PROPELLER DWI with identical acquisition matrix (128 x 128) demonstrates few artifacts and provides significantly better image quality In this "Turbo" implementation of PROPELLER, five gradient echoes were acquired between each refocusing RF pulse to speed up the acquisition. The overall acquisition adds approximately 100 s to the scan time.

Kspace Propeller Blade
Fig. 3.56. Very high-resolution (512 x 512) DTI results acquired with a self-navigated variable-density spiral twice-refocused spin-echo (TRSE) sequence (SNAILS). (a) T2w-SNAIL image (b = 0).

(b) Trace (isotropic average of x, y, and z) diffusion-weighted image. (c) FA map. Although the scan times are prohibitively long, this demonstrates the potential of self-navigated multi-shot DWI methods.

B1000 Sense Diffusion And Artefacts

Fig. 3.57. (128 x 128, b = 1000) SE-EPI DWI remove white matter anisotropy effects), and single-shot 128 x 128 images from four (left ADC "expo" images are shown from top to to right) of 28 slices acquired in 24 s along the bottom at left. The ADC "expo" maps are x-, y-, and z-gradient directions. The b0, DWI, processed to remove long ADC values such as trace ADC ("averaged" along three axes to CSF, which typically hides lower ADC values.

Fig. 3.57. (128 x 128, b = 1000) SE-EPI DWI remove white matter anisotropy effects), and single-shot 128 x 128 images from four (left ADC "expo" images are shown from top to to right) of 28 slices acquired in 24 s along the bottom at left. The ADC "expo" maps are x-, y-, and z-gradient directions. The b0, DWI, processed to remove long ADC values such as trace ADC ("averaged" along three axes to CSF, which typically hides lower ADC values.

Was this article helpful?

0 0

Responses

  • Genet
    What is subcortical stroke?
    7 years ago

Post a comment